1990s 1993 Seville AC Problem

[Mike Shawgo]

Hi everyone,

I have a 1993 Seville with 77,000 miles, and a friend in another city has a 1992 Seville with I think around 90,000 miles. We are BOTH having the identical issue with our AC systems. Both were converted years ago to r134a. I've had a mechanic recharge mine, seems to be working when I pick it up.  The next day I get the message "Very Low Refrigerant, Compressor Off" and an error code A047. I have replaced the low pressure sensor, the mechanic has replaced the condenser and the accumulator, still no luck. The mechanic has put dye in the freon to detect a leak but has found nothing. The compressor, during the short time it runs, seems to be fine, doesn't make any noise. I ran across something which said that many mechanics use a conversion which uses less r134a refrigerant than would be called for with R12 which could trigger the low refrigerant code, but my mechanic assured me that he used the same amount as the original R12 amount. Mechanic still thinks it's something electrical, such as a sensor, but I don't know what it could be.  Any ideas anyone?

Mike Shawgo
#20545

["Cadillac Kid" Greg Surfas 15364]

What is anyone doing to leak check the system. If the low refrigerant code didn't appear while the system was working then hence there is a leak somewhere. Check the evaporator as well
Greg Surfas

[TJ Hopland]

So when they re check the system is it actually reading low?

[Mike Shawgo]

TJ, no I don't think it is reading low. The mechanic keeps checking wiring, sensors, etc. I keep thinking it might be the compressor, but usually there are other symptoms of a bad compressor. The low refrigerant message has appeared just a matter of hours after I pick it up from the repair shop. I thought about the evaporator as well, but isn't that a real pain to get to?

[Mike Shawgo]

I guess the other thing I keep thinking about is that my friend's 92 Seville is doing exactly the same thing--they charge it up, and the next day he is getting the low refrigerant compressor off message.

[James Landi]

Many areas in the system for leaks.  If the technician verifies that your system is, indeed, low on refrigerant after you receive the message, then there is a leak... slow, hard to detect, yet causing the error message. Has the tech put in black light sensitive dye? This dye can create a telltale leak trail, however, tiny leaks in the evaporator that, as you know, is entirely hidden in the evaporator air box are hard to detect.  If you're fortunate, "o" rings or even a failing schrader valve can create tiny, hard to locate refrigerant leaks. I suspect that the co-incidence with your friend's Cadillac is simply a fact that we've all faced, 30 year old cars develop failures due to age, and are so very challenging, frustrating, and costly to repair. 

[TJ Hopland]

Do we know if 92 and 93 are more or less the same system?  93 was the Northstar?  So at least a different engine computer but perhaps the body hvac one is the same?    And you say your friend's gets charged so perhaps his is actually a leak?

Do you have a service manual?  If not I think that would be the next step.  Someone needs to read the circuit description and see what sensors are involved and what the thresholds are for tripping the code. Also would be good to know can it self reset or once its tripped does it stay that way.  Once you know that stuff maybe the next step would be to hook up the gauges and run it to see if pressures are getting anywhere near the trip thresholds.  Its possible that the 134 differences are actually the problem with this monitoring system.   There could just be a momentary blip in the pressures that with R12 didn't happen but happens with 134.  If that is the problem not sure what can be done, try to fool the sensor calibration?   Mess with the gas volume to see if that changes the blip without harming performance?

If the pressures are no where near the thresholds then you have to be looking for an electrical problem and the most common one would be grounds.  Get the electrical diagram and see where the ground points are for the sensors and what ever computer is running that stuff.  Find those points and take em apart for a good cleaning and inspection. 

[Mike Shawgo]

No the 1993 Seville has the 4.9 engine, same as the 92. I have the shop manual, but I'm not working on it myself. That's good idea about the circuits connected to the low pressure sensor. Although my mechanic did have an electrical diagram so he may have already gone down that route. I'm thinking the leaking theory is probably correct, although if it leaks enough in one day to trigger the low pressure sensor, it seems like it wouldn't be that difficult to find.  Thanks guys for your help!

Mike

[TJ Hopland]

The manual should have information that says something like code is set when pressure on S1 is below 20 psi for more than 3 seconds and while R3 is engaged.  It should also tell you what and where S1 and R3 are and where.  There should also be a section that explains the overall operations and what controls what and when.  Maybe R3 is for a fan and thats a clue?  Maybe the fan isn't running when it should.   I just made all that up as an example of what the manual should say so don't go looking for S1 and R3.       

[dn010]

my mechanic assured me that he used the same amount as the original R12 amount.

Using the "same amount" is incorrect. You don't need as much R134 in a system designed for R12. If he is putting in the same amount, then it is being overfilled. R134 operates at a higher pressure to begin with so using the same amount would have you running at even higher pressure which could certainly lead to leaks. Because it has a higher operating pressure, using less should not trigger a low pressure warning although you may have to adjust (if possible) the low pressure switch to have it cycle more efficiently. To have it filled properly, the mechanic needs to figure out how much R12 is originally supposed to go in and then calculate how much R134 should be replacing it. There is a formula online somewhere. Second problem is the R12 hoses are not designed for R134 which require barrier hose which means your R12 hoses, while it may take a long time, are allowing R134 to escape.
For what it's worth, the most difficult leak I've found was a leaking front seal on a compressor. I checked everything else and couldn't figure it out and just happened to notice the front seal (not on a cadillac). Other issues I've found were bad schrader valves or ones that were not screwed all the way down. Last leak I fixed was an o-ring that was too big for a high pressure switch, the ring could not compress enough and squished down and away from the switch causing a leak, of course only when the AC would run (higher pressure) and not leak once the compressor stopped and the pressure stabilized.

AC is not an easy thing, if it is a regular mechanic doing the work, might be worth it to take it to an actual AC shop that specialize in this sort of thing.

[Mike Shawgo]

Dan,

Makes sense, although I have run across several discussion groups (Cadillac Forum) where this is discussed, and they swear up and down that when they used the same amount of r134a as r12 (instead of the conversion formula) they stopped getting the low pressure code. But your explanation makes sense. Also I think you are right that I need to take it to a shop which specializes in AC. Thanks Dan.

--Mike

["Cadillac Kid" Greg Surfas 15364]

Finding the "probable" leak here is the issue. Checking for evaporator leaks is simply a matter of (with pressure in the system) running the fan and sticking the leak detector probe into one of the air outlets in the car.
The issue of r-134a vs R-12 has lead to many,many urban myths. First of all, yes R-134a has a higher CONDENSING PRESSURE for a given condensing temperature, meaning the high side operates at higher pressures. You would think that this would force more refrigerant through the flow control device (TX valve, oirfice, etc.),but since R-134a is more dense than R-12 the flow dynamics slow down the flow at that point.  Low side pressures of the two refrigerants through an R-12 system are just about equal, and at those pressures, R-134a is just about 18% LESS dense than R-12, meaning fewer pounds per cubic foot of vapor.  The R-12 compressor (A-6, R-4) is a positive displacement unit, meaning it "pumps" X number of cubic feet per revolution and this equates to about 18% less volume of 134a, hense the reason for the drop in capacity when charging an R-12 system with R-134a.
As far as the reason for a 10% lesser charge (by pounds) of R-134a, in the liquid phase, R-134a is 8.4% more dense than R-12, meaning the same amount by weight will take up more space. Charging to R-12 weight specs results in liquid backing up in the condenser, essentially reducing condensing space and again raising pressures.
Now, find the leak and it sounds like you may have solved this issue.
Greg Surfas

[Mike Shawgo]

Hi guys,

Well I finally found a Radiator and Air Conditioning repair shop that was highly recommended by another mechanic. He did a performance test with the dye and infrared light detector. He said there was a minor leak from a low-side Schrader valve, which he replaced. He replaced the orifice tube and flushed the condenser and lines, then recharged with the correct amount indicated for R134a refrigerant. It's working great so far.  He told me to drive it this week with the AC on, then bring it back in next week and he will recheck with the light for any other leaks. This whole procedure just took a few hours. I also informed my friend with the 92 Seville that he needed to seek out an AC specialist, not just a general mechanic, to check his system for leaks, and they should probably do the same thing--flush the system and replace the orifice tube. He may need a new condenser and accumulator as well, since mine had already been replaced. So, you guys were right!  Thanks so much!

Mike